A Greedy Merge Learning Algorithm for Gaussian Mixture Model
Yan Li, Lei Li · 2009
Gaussian mixture model (GMM) has been widely used in fields of image processing and investment data mining. However, in many practical applications, the number of the components is not known. This paper proposes a kind of greedy merge EM (GMEM) learning algorithm such that the number of Gaussians can be determined automatically with the minimum message length (MML) criterion. Moreover, the greedy merge learning algorithm is successfully applied to unsupervised data analysis. It is demonstrated well by the experiments that the proposed greedy merge EM (GMEM) learning algorithm can make both parameter learning and decide the number of the Gaussian mixture.